1dnl  ARM mpn_mod_34lsub1 -- remainder modulo 2^24-1.
2
3dnl  Copyright 2012, 2013, 2018 Free Software Foundation, Inc.
4
5dnl  This file is part of the GNU MP Library.
6dnl
7dnl  The GNU MP Library is free software; you can redistribute it and/or modify
8dnl  it under the terms of either:
9dnl
10dnl    * the GNU Lesser General Public License as published by the Free
11dnl      Software Foundation; either version 3 of the License, or (at your
12dnl      option) any later version.
13dnl
14dnl  or
15dnl
16dnl    * the GNU General Public License as published by the Free Software
17dnl      Foundation; either version 2 of the License, or (at your option) any
18dnl      later version.
19dnl
20dnl  or both in parallel, as here.
21dnl
22dnl  The GNU MP Library is distributed in the hope that it will be useful, but
23dnl  WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
24dnl  or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
25dnl  for more details.
26dnl
27dnl  You should have received copies of the GNU General Public License and the
28dnl  GNU Lesser General Public License along with the GNU MP Library.  If not,
29dnl  see https://www.gnu.org/licenses/.
30
31include(`../config.m4')
32
33C	     cycles/limb
34C StrongARM	 ?
35C XScale	 ?
36C Cortex-A5	 2.67
37C Cortex-A7	 2.37
38C Cortex-A8	 2.34
39C Cortex-A9	 ?
40C Cortex-A15	 1.39
41C Cortex-A17	 1.60
42C Cortex-A53	 2.51
43
44define(`ap',	r0)
45define(`n',	r1)
46
47C mp_limb_t mpn_mod_34lsub1 (mp_srcptr up, mp_size_t n)
48
49C TODO
50C  * Write cleverer summation code.
51C  * Consider loading 6 64-bit aligned registers at a time, to approach 1 c/l.
52
53ASM_START()
54	TEXT
55	ALIGN(32)
56PROLOGUE(mpn_mod_34lsub1)
57	push	{ r4, r5, r6, r7 }
58
59	subs	n, n, #3
60	mov	r7, #0
61	blt	L(le2)			C n <= 2
62
63	ldmia	ap!, { r2, r3, r12 }
64	subs	n, n, #3
65	blt	L(sum)			C n <= 5
66	mov	r7, #0
67	b	L(mid)
68
69L(top):	adds	r2, r2, r4
70	adcs	r3, r3, r5
71	adcs	r12, r12, r6
72	adc	r7, r7, #0
73L(mid):	ldmia	ap!, { r4, r5, r6 }
74	subs	n, n, #3
75	bpl	L(top)
76
77	adds	r2, r2, r4
78	adcs	r3, r3, r5
79	adcs	r12, r12, r6
80	adc	r7, r7, #0		C r7 <= 1
81
82L(sum):	cmn	n, #2
83	movlo	r4, #0
84	ldrhs	r4, [ap], #4
85	movls	r5, #0
86	ldrhi	r5, [ap], #4
87
88	adds	r2, r2, r4
89	adcs	r3, r3, r5
90	adcs	r12, r12, #0
91	adc	r7, r7, #0		C r7 <= 2
92
93L(sum2):
94	bic	r0, r2, #0xff000000
95	add	r0, r0, r2, lsr #24
96	add	r0, r0, r7
97
98	mov	r7, r3, lsl #8
99	bic	r2, r7, #0xff000000
100	add	r0, r0, r2
101	add	r0, r0, r3, lsr #16
102
103	mov	r2, r12, lsl #16
104	bic	r1, r2, #0xff000000
105	add	r0, r0, r1
106	add	r0, r0, r12, lsr #8
107
108	pop	{ r4, r5, r6, r7 }
109	return	lr
110
111L(le2):	cmn	n, #1
112	bne	L(1)
113	ldmia	ap!, { r2, r3 }
114	mov	r12, #0
115	b	L(sum2)
116L(1):	ldr	r2, [ap]
117	bic	r0, r2, #0xff000000
118	add	r0, r0, r2, lsr #24
119	pop	{ r4, r5, r6, r7 }
120	return	lr
121EPILOGUE()
122